Spin crossover composite materials for

electrothermomechanical actuators
Il’Ya A. Gural’Skiy, Carlos Mario Quintero, José Sánchez Costa, Philippe
Demont, Gabor Molnar, Lionel Salmon, Helena J. Shepherd, Azzedine
Bousseksou

To cite this version:

Il’Ya A. Gural’Skiy, Carlos Mario Quintero, José Sánchez Costa, Philippe Demont, Gabor Mol-
nar, et al.. Spin crossover composite materials for electrothermomechanical actuators. Journal
of Materials Chemistry C, Royal Society of Chemistry, 2014, vol. 2 (n 16), pp. 2949-2955.
<10.1039/c4tc00267a>. <hal-01166077>

HAL Id: hal-01166077

https://hal.archives-ouvertes.fr/hal-01166077
Submitted on 22 Jun 2015

HAL is a multi-disciplinary open access L’archive ouverte pluridisciplinaire HAL, est

archive for the deposit and dissemination of sci-
entific research documents, whether they are pub-
lished or not. The documents may come from
teaching and research institutions in France or
abroad, or from public or private research centers.
destinée au dépôt et à la diffusion de documents
scientifiques de niveau recherche, publiés ou non,
émanant des établissements d’enseignement et de
recherche français ou étrangers, des laboratoires
publics ou privés.
Open Archive TOULOUSE Archive Ouverte (OATAO)
OATAO is an open access repository that collects the work of Toulouse researchers and
makes it freely available over the web where possible.

This is an author-deposited version published in : http://oatao.univ-toulouse.fr/

Eprints ID : 11557

To link to this article : doi: 10.1039/c4tc00267a

URL : http://dx.doi.org/10.1039/c4tc00267a

To cite this version : Gural'skiy, Il'ya A. and Quintero, Carlos M. and

Costa, José Sánchez and Demont, Philippe and Molnar, Gabor and
Salmon, Lionel and Shepherd, Helena J. and Bousseksou, Azzedine
Spin crossover composite materials for electrothermomechanical
actuators. (2014) Journal of Materials Chemistry. C, vol. 2 (n° 16). pp.
2949-2955. ISSN 2050-7526

Any correspondance concerning this service should be sent to the repository

administrator: staff-oatao@listes-diff.inp-toulouse.fr
DOI: 10.1039/c4tc00267a

Spin crossover composite materials for

electrothermomechanical actuators†
Il'ya A. Gural'skiy,ab Carlos M. Quintero,c José Sánchez Costa,a Philippe Demont,d
Gábor Molnár,a Lionel Salmon,a Helena J. Shepherd*e and Azzedine Bousseksou*a

Composites of the spin crossover complex [Fe(trz)(H-trz)2](BF4) (H-trz ¼ 1,2,4-4H-triazole and trz ¼ 1,2,4-
triazolato) dispersed in a poly(methylmethacrylate) (PMMA) matrix were synthesized and investigated for
their spin crossover properties by optical reflectivity, Raman spectroscopy and calorimetry. These
composite films were used to fabricate bilayer cantilevers that can perform efficient and tuneable
mechanical actuation based on the spin transition. A prototype device that uses the spin transition
phenomenon to convert electrical energy into mechanical motion through Joule heating is described.
This device is used to perform oscillatory actuation driven by a modulated current. The ability to tune the
performance of this electromechanical system is demonstrated by varying the working temperature, the
applied ac current and its frequency.

Introduction
The transformation of electrical energy into motion is one of the
main demands from micro- and nanoelectromechanical
systems (MEMS and NEMS).1 This conversion could be achieved
via electrostatic2 or piezoelectric effects,3 Joule heating,4 etc.
Among these methods electrothermal actuation is a widely used
approach due to its simplicity both for the fabrication process
and the actuation mechanism.5 The choice of the actuating
material is a crucial step in the construction of a new electro-
mechanical system and the numerous specic requirements
make the development of novel actuating materials necessary.6
For instance, the energy density of a mechanical actuator is
basically dependent on the strain that can be generated in its
active materials. However, many other important performance
indicators, such as the displacement, the force, actuation speed
(or frequency) and energy efficiency will also depend on the
material selection.
Spin crossover (SCO) complexes of 3d4–3d7 transition metal
ions are well known for changing a variety of physical properties
a
LCC, CNRS & University of Toulouse (UPS, INPT), 205 route de Narbonne, 31077
Toulouse, France. E-mail: azzedine.bousseksou@lcc-toulouse.fr
b
Department of Chemistry, Taras Shevchenko National University of Kyiv, 64
Volodymyrska Street, 01601 Kyiv, Ukraine
c
LAAS, CNRS & Université de Toulouse (UPS, INSA, IAES), 7 avenue du colonel Roche,
31077 Toulouse, France
d
Institut Carnot CIRIMAT, CNRS & Université de Toulouse III, 118 Route de Narbonne,
31062 Toulouse, France
e
Department of Chemistry, University of Bath, Bath, BA2 7AY, UK. E-mail: h.j.
shepherd@bath.ac.uk
† Electronic supplementary information (ESI) available: Actuation of a cantilever
with a different geometry, detailed theoretical calculations, movie of the cantilever
oscillation. See DOI: 10.1039/c4tc00267a
during the transition between their high spin (HS) and low spin
(LS) electronic congurations.7,8 Despite copious investigations
into their magnetic and optical properties, the spontaneous
linear strain (3) that can reach 8% (ref. 9 and 10) in these
materials seems to be one of the most fascinating assets for
practical applications.11 The 3 values in SCO complexes are high
when compared to analogous phase transition materials. For
example, in the very promising VO2 known for the reversible
metal–insulator phase change, a linear strain up to 2% can be
reached.12–16 Another important asset is that SCO can be
provoked by different external stimuli such as temperature,
pressure, magnetic eld, excitation with light or adsorption of
chemical species. The interest is additionally supported by the
existence of hundreds of SCO complexes reported to date with
different transition temperatures, hysteresis and abruptness of
the spin transition.7 This versatility can be further extended by
the elaboration of composite materials. For example, SCO
compounds have been imbedded in silica17–19 and various
polymer20–25 matrices providing nanoscale connement and/or
better processability.
Recently we reported on the application of spin crossover
materials for the elaboration of bimorph cantilevers, which
were actuated both thermally and by light irradiation.11 As a
proof of concept, a single crystal of the SCO complex
{Fe(3-CNpy)[Au(CN)2]2}$2/3H2O (py ¼ pyridine) was covered
with a layer of Al and used as a cantilever actuator. A reversible
movement of this bimorph was reached as a result of thermally
and photo-induced spin transitions, thus revealing a novel
mechanism of actuation that opens up perspectives for the
development of MEMS, articial muscles and other applica-
tions for the harvesting and transformation of energy (thermal,
electric, photonic, .). Additionally, three bilayer polymer/SCO
composite devices, each with different SCO compounds, were
developed to demonstrate the versatility of SCO materials for
actuating tasks. In this paper we go further by describing a
prototype device that can exploit the spin transition phenom-
enon to convert electrical energy into motion via Joule heating
in a composite bilayer actuator.
Results and discussion
SCO/polymer composites
Even though the application of a single crystal SCO material for
the construction of bimorph cantilevers was found efficient and
the behavior of these actuators is easy to rationalize from a
theoretical point of view, the practical use of such systems has
certain limitations. This is mostly related to the serendipity of
the crystal growth. In contrast, techniques for employing poly-
mers and polymer composites in MEMS design are well devel-
oped.26 Hence, we synthesized a polymer composite material,
consisting of the spin crossover complex [Fe(trz)(H-trz)2](BF4)
(form 1b, one of the two forms described by Kröber et al.27) and
the organic polymer PMMA, which can produce continuous
lms of several microns in thickness over a large area. This
iron(II) compound is known for its abrupt and robust spin
transition that takes place with a large thermal hysteresis
around 85  C on heating and 60  C on cooling. It is important to
note also that in this particular SCO complex the transition is
accompanied by a volume change of 11%.28
A series of PMMA/1b composites with different relative
proportions of the active complex were fabricated by mechan-
ical dispersion of the powder of 1b in a PMMA solution with
subsequent evaporation of the chloroform solvent. A drop-
casting procedure allows the deposition of the composite lms
over different surfaces. Films with different proportions (10–
50%) of the active complex were produced (Fig. 1). Loading in
excess of 50% of the SCO complex results in lms that are brittle
Fig. 1 Temperature dependence of the optical reflectance (l ¼ 550
nm) of [Fe(trz)(H-trz)2](BF4) (100%) and of the composite materials
PMMA/[Fe(trz)(H-trz)2](BF4) with different loadings (10–50%) of the
active complex. Heating and cooling modes are indicated by arrows.
The inset shows the photographs of the composite films.
and cannot be removed from the substrate or further processed.
The spin transition properties of these composites were
observed and analyzed by optical microscopy since the SCO is
accompanied by a drastic change of color in the active complex
from pink in the LS state to colorless in the HS state. This effect
is associated with the reversible bleaching of the characteristic
1
A1 / 1T1 absorption band centered around l ¼ 550 nm. Fig. 1
reveals a strong thermal dependence of the composite color that
is clearly associated with the HS 4 LS spin transition. The SCO
in the composites takes place within the same region of
temperature as for the pure complex. The transition curves of
the composites were found to be relatively independent of the
polymer/complex ratio; the most signicant differences being a
slight loss of abruptness of the spin transition for composites
with lower loadings of the active material. This observation may
simply be related to the lower thermal conductivity of the
polymer component compared to the coordination compound.
While the SCO behaviour of composite materials is in general
not easy to predict, the negligible inuence of a polymer matrix
on those properties, as shown here, is certainly not
unprecedented.20–23
Selected Raman spectra of [Fe(trz)(H-trz)2](BF4) and the
PMMA/[Fe(trz)(H-trz)2](BF4) composite (50% loading) recorded
at 20  C and 110  C are shown in Fig. 2. In the low frequency
region (100–300 cm1), very signicant changes are observed
upon the SCO. As discussed in ref. 29, the modes at 300, 288,
211, 202 cm1 are attributed to the LS state of the complex,
whereas those at 186 and 109 cm1 are associated with the HS
state. These changes of the vibrational frequencies accompa-
nying the spin transition are a general trend and originate from
the strong electron–lattice coupling in these systems.7 The
disappearance of the characteristic LS/HS modes at high/low
temperatures allows us to conclude that the spin transition is
virtually complete in both directions for the isolated complex as
well as for the composite material.
Fig. 2 Variable-temperature Raman spectra of the [Fe(trz)(H-
trz)2](BF4) complex and the PMMA/[Fe(trz)(H-trz)2](BF4) (50%)
composite material.
 Differential scanning calorimetry (DSC) curves of [Fe(trz)(H-
trz)2](BF4) and the PMMA/[Fe(trz)(H-trz)2](BF4) composite (50%)
are shown in Fig. 3.
In agreement with the optical reectivity and Raman spec-
troscopy measurements, these data demonstrate a high degree
of similarity between the SCO in the isolated complex and the
composite materials. Shis of the transition temperatures
between the complex and its composite are attributed to ther-
malization issues – an organic polymer would be expected to be
less thermally conductive than a coordination compound. This
measurement allows also the calculation of the enthalpy (DH ¼
25 kJ mol1) and entropy (DS ¼ 73 J mol1 K1) changes asso-
ciated with the spin transition in 1b. These values were found
similar for the isolated complex and the complex embedded in
PMMA and they are also in good agreement with the data
reported in ref. 27 for the pure complex.
Bilayer actuators
The composite lm with the highest proportion of active
material (50%) was then used to construct bilayer actuators. The
highest possible loading by the active compound is expected to
optimize the strain that can be transmitted from the complex to
the polymer matrix, thus maximizing the actuating capabilities.
Despite the proportion of the strain absorbed by the exible
polymer matrix, the spin transition in the active component
results in a signicant isotropic expansion of the composite
material.
In a simple bilayer strip, the difference in expansion of the
two layers results in the deection of the cantilever that can be
exploited for actuation. In our device, the second (inert) layer is
used not only to generate the strain mismatch, but is also used
as a heating element via the Joule effect. In such a way, the
application of a current to the cantilever should provoke a spin
transition (from LS to HS), while removing the current should
cause a dissipation of heat followed by the reverse transition
(HS to LS). We selected a polymer that was lled with silver
Fig. 3 DSC curves of the pure [Fe(trz)(H-trz)2](BF4) complex and the
PMMA/[Fe(trz)(H-trz)2](BF4) (50%) composite material acquired at
heating/cooling rates of 10  C min1.
Fig. 4 Schematic representation of a SCO bilayer cantilever with
electrothermal actuation.
particles (industrially available conductive paint) to act as the
conductive layer because this conductive composite has a much
higher adhesion to the active layer than a simple metal layer can
have. Moreover, this choice for the conductive component
makes both layers of the bimorph similar in terms of thermal
expansion and Young's modulus. This minimizes the mismatch
of the thermal expansion between the two layers and maximizes
the deection of the strip (for a given geometry).
The bilayer lms were fabricated by spin-coating the
conductive composite solution over a PMMA/SCO composite
lm. A cantilever was subsequently cut from this bilayer and
connected to two electrodes: one electrode is xed, whereas the
second electrode is made of a exible thin aluminium foil so as
to allow (nearly) free movement with the cantilever (Fig. 4).
This setup was placed in a cryostat to control the tempera-
ture around the cantilever. The behavior of this cantilever was
initially studied within a range of 23–100  C. Heating was
stopped at 100  C to avoid any transformations of the polymers
(e.g. glass transition) and damage of the device. As temperature
increases the spin transition in [Fe(trz)(H-trz)2](BF4) causes a
spontaneous strain that results in a strong deformation of the
bimorph at the temperature of the spin transition. The occur-
rence of SCO can also be clearly inferred from the simultaneous
change of the composite color during the actuation (Fig. 5). The
movement of the cantilever was quantied by following with a
video-microscope the position of the end connected to the
exible electrode. Displacement as a function of temperature is
shown in Fig. 5a.
The hysteresis loop of the thermal actuation (Fig. 5a) is
rather deformed in comparison with the spin transition loop
(Fig. 1). This is attributed to inefficient heat transfer from the
heating plate to the cantilever suspended in air, attached only at
the ends. In addition to heat exchange effects, mechanical
stress can also inuence the form of this curve. However, the
limited region of temperature as well as the hysteretic nature of
the deection clearly demonstrate that tip deection is the
result of the spin transition in the active complex.
The amplitude of actuation (ca. 1600 mm) allows us to
calculate the reactive force at the end of the cantilever and the
effective linear strain of the composite material to be 20 mN
and 1%, respectively (see ESI† for details on the calculations).
These values are slightly underestimated since the exible
electrode acts as a load on the cantilever. Nevertheless the
 Fig. 6 Electric current induced actuation of a bimorph cantilever. Tip
position of the cantilever as a function of time (upper panel) and the
corresponding applied current signal (bottom panel). The frequency of
the alternative current is 0.1 Hz, Imax ¼ 300 mA, T ¼ 20  C.

additional load, it does not play a crucial role in the movement

Fig. 5 Thermal actuation of a bimorph cantilever. (a) Tip position of
the cantilever as a function of temperature. Heating and cooling
modes are indicated by arrows. (b) Photographs (8.6 mm  7.6 mm) of
the cantilever at 68  C in the cooling and heating modes demonstrate
a shape memory effect, which occurs due to the thermal hysteresis of
the spin transition. The color change associated with the spin transition
is also observed (see also ESI Movie SM1†).
effective strain is signicantly lower than that estimated from
the additive effect (ca. 2%), which we relate to the partial
absorption of strain by the polymeric matrix. The efficiency of
this material for actuation is described by the volumetric work
density W/V, i.e. the maximal mechanical output, which can be
produced by a given volume of the composite. W/V for the 50%
composite was calculated to be ca. 150 mJ cm3 (see ESI† for
details on the calculations). This value is comparable or higher
than those reported for piezo-electric, magnetostrictive and
electroactive polymer materials.30
of the cantilever. This is due to the force (20 mN) generated at
the tip of the cantilever being considerably higher than the force
necessary to move the aluminium foil.
Aer an initial “run-in” period of a few cycles, the actuating
amplitude becomes highly repeatable within the given oper-
ating conditions (Fig. 6). We observed that this amplitude
depends strongly on the experimental conditions. The most
important contributing factor is the fraction of the complex that
undergoes SCO when the current is turned on/off. This leads to
different amplitudes of deection with different applied
frequency, current and temperature of the surrounding
medium. A series of experiments were designed to probe the
interdependence of these three parameters (Imax, f and T).
The amplitude of cantilever movement is shown as a func-
tion of applied current in Fig. 7. A current below 200 mA does

Electrothermal actuation of the cantilever

To generate an electrothermally driven actuation, a squared
current (I) was applied to the cantilever, which engenders Joule
heating. In a typical experiment, I was changed stepwise from
0 to Imax and back to 0 at a constant frequency (f). When I ¼ Imax,
the LS / HS transition is induced, whereas removing the
current leads to the backward HS / LS transition. This
sequence of LS / HS / LS transitions provokes oscillating
motion of the cantilever (Fig. 6). The amplitude of this move-
ment is constant in numerous cycles (see also ESI Movie SM1†).
Fig. 7 Amplitude of oscillation of the bilayer cantilever as a function of
In this experiment, the device was passed through more than the applied current Imax at constant temperature and frequency (T ¼
300 actuating cycles with no observed fatigue of the actuator. It 20  C, f ¼ 0.1 Hz). Insets: tip position of the cantilever as a function of
is worth noting that though the exible electrode acts as an time for a few selected values of Imax.

Fig. 8 Amplitude of the bilayer cantilever oscillation at different
temperatures. The ac current and frequency were fixed (Imax ¼ 300
mA, f ¼ 0.1 Hz). Insets: tip position of the cantilever as a function of
time at selected temperatures.
not result in any signicant actuation of the cantilever. This
threshold effect is likely to be the result of insufficient heating,
i.e. below a certain temperature the nucleation of the HS phase
cannot take place. The low amplitude oscillation observed for
currents below this value should be considered chiey as a
thermal expansion effect without SCO. By contrast, a current over
200 mA generates enough heat to induce the thermal SCO,
reaching a maximum amplitude of actuation at Imax z 300 mA.
The deection at currents between 200 and 300 mA corresponds
to a partial spin transition of the Fe(II) complex in the composite.
This threshold current value can be tuned via several different
approaches, including the change of SCO active complex. Indeed,
the use of any SCO compound with higher/lower temperature of
transition should increase/decrease the critical value of current
necessary to drive the cantilever oscillation.
Measurements at constant f and Imax reveal a clear dependence
of the actuation amplitude on the surrounding temperature as
shown in Fig. 8. Up to 50  C a slight and almost linear diminution
of the actuating amplitude is observed and can be explained by an
incomplete HS to LS transition when the current is switched off.
Above 50  C an abrupt drop off in amplitude is observed, due to
the thermal SCO occurring above this point. Thus, even when
there is no current applied to the cantilever, a signicant fraction
of the active SCO complex remains in the HS state.
Again, this characteristic cut-off temperature can be tuned by
substituting the active complex for another SCO material. The
abrupt on/off switch of the actuation with temperature can be
achieved by using highly cooperative spin crossover systems
where the spin state change occurs sharply within < 1 K. (N.B. To
observe an abrupt cut-off as a function of temperature efficient
thermalization is also necessary.) On the other hand, the use of
compounds with a gradual SCO – taking place within hundreds
degrees – can allow ne control over the actuation amplitude by
changing the environmental temperature.
Varying the frequency at constant Imax and T reveals a cut-off
of the actuation (dened as a decay in amplitude of 0.707 of the
Fig. 9 Amplitude of the bilayer strip oscillation as a function of
frequency at constant temperature and ac current (T ¼ 20  C, Imax ¼
300 mA). Insets: tip position of the cantilever as a function of time at
selected frequencies.
maximum value) at fcut off z 0.5 Hz (Fig. 9). In this case the
completeness of the transition in both directions (LS to HS and
HS to LS) is limited by the rate of heat diffusion; there is
insufficient time for the device to reach the required tempera-
ture to switch all active centers. For this actuator there is no
frequency limitation up to f ¼ 0.1 Hz; however, no movement is
observed for f > 1 Hz. This limitation is clearly related to the heat
transfer processes and not to the spin conversion process that
has much faster dynamics.7,8
The observed cut-off frequency is not a fundamental limit, but
rather denotes that (if necessary) the device must be operated at a
different temperature and/or a different current should be
applied. Even more importantly, the cantilever geometry and the
thermalisation conditions can also drastically change the
performance of these actuators in terms of actuating speed. One
example describing the effect of a different geometry is shown in
the ESI (Fig. S1–S3†). A cantilever of a smaller size was fabricated
and studied in a similar manner. A direct consequence of the size
reduction is the faster heat exchange between the cantilever and
its surroundings. Thus, a lower cut-off temperature and current,
as well as a higher cut-off frequency were observed.
Experimental information
PMMA (Mw ¼ 550 000), iron(II) tetrauoroborate hexahydrate,
1-H-1,2,4-triazole and chloroform were purchased from Sigma
Aldrich and used without any further purication. Conductive
silver paint RS 186-3593 was purchased from RS Components.
This paint contains a polymeric basis lled with silver micro-
spheres (<18 mm, 56%) and dissolved in an n-butyl acetate/
2-methoxy-1-methylethyl acetate mixture.
[Fe(trz)(H-trz)2](BF4) (form 1b) was prepared from
Fe(BF4)2$6H2O and 1-H-1,2,4-triazole according to the procedure
described by Kröber et al.27 The active composite material was
elaborated by mechanical dispersion of the [Fe(trz)(H-trz)2](BF4)
powder (5 mg) in a solution of PMMA (5 mg) in CHCl3 (200 ml).
This suspension was drop-casted onto a silicon substrate (1  1
cm2) and allowed to dry. 150 ml of the conductive paint was
subsequently spin coated (2000 rpm, 3000 rpss, 30 s) onto the
upper surface. The resulting bilayer lm was removed from the
substrate and cut into cantilevers. The cantilever used was 6.7
mm in length and 1 mm in width. The thicknesses of the SCO
and conductive layers were ca. 100 and 50 mm, respectively.
Variable temperature Raman spectra of the samples were
acquired by means of an Xplora (Horiba) microspectrometer
and a Linkam THMS600 cryostat. A long working-distance
objective (50, numerical aperture ¼ 0.5) was used to focus the
laser beam (638 nm, 0.1 mW) on the sample and to collect the
scattered photons. Rayleigh scattering was removed by an edge
lter. Differential scanning calorimetry analysis was carried out
on a Netsch DSC 204 instrument under helium purging gas. The
measurements were conducted between 0 and 100  C at a scan
rate of 10 K min1 both on heating and cooling. Temperature
and enthalpy were calibrated using the melting transition of
standard materials (Hg, In, Sn).
In order to perform the electrical actuation experiments, a
cantilever was xed onto a xed steel electrode with the
conductive paint. The other side was attached to a exible
aluminium electrode (10 mm thick) in the same manner. The
temperature was controlled using a Linkam HFS600E cryostat
equipped with exible internal microelectrodes to connect the
device to the current source. The videomicroscopy and optical
reectivity images were acquired with a MOTIC stereomicro-
scope equipped with a colour camera. Current pulses were
generated with a programmable current source (Keithley 6430).
Tip displacement was determined from the video images. The
image treatments were carried out under ImageJ and MatLab
environments as described previously.11
Conclusions and perspectives
The spin crossover complex [Fe(trz)(H-trz)2](BF4) was dispersed in
an organic polymer matrix to form freestanding lms. The spin
crossover properties of the isolated complex were preserved in the
composites, which allowed us the development of a highly effi-
cient electromechanical actuating device with a volumetric work
density of ca. 150 mJ cm3. The prototype device described here
exhibited repeatable and tunable actuation via temperature,
frequency and current variation and provided both large force
(several mN) and large displacement (several mm). The procedure
to elaborate the active composites can be easily extended not only
to other SCO compounds, but also to different phase transition
materials (VO2, etc.). This approach of actuation driven by SCO can
be also introduced in MEMS/NEMS sensors, since several SCO
compounds have shown a good sensitivity to different hazardous
compounds.31 An exciting perspective of this work will be the
investigation of the effects of scaling, since the faster heat transfer
rate at reduced scales will result in higher work frequency and
effectiveness.5 This prospect is tangible since the spin transition
properties can be preserved down to very small sizes.8
Acknowledgements
Financial support from the project Switchelec (ANR-12-IS07-
0003-01) is acknowledged. JSC thanks the Marie-Curie research
program (NanoSCOpe 328078).
Notes and references
1 M. Bao, in Analysis and Design Principles of MEMS Devices,
Elsevier Science, Amsterdam, 2005, pp. 1–32.
2 Comprehensive Microsystems, ed. Y. Gianchandani, O. Tabata
and H. Zappe, Elsevier, Oxford, 2008, vol. 2, pp. 1–100.
3 T.-L. Ren, H.-J. Zhao, L.-T. Liu and Z.-J. Li, Mater. Sci. Eng., B,
2003, 99, 159–163.
4 S. Pal and H. Xie, J. Micromech. Microeng., 2012, 22, 115036.
5 P. Srinivasan and S. M. Spearing, J. Microelectromech. Syst.,
2008, 17, 653–667.
6 MEMS Materials and Processes Handbook, ed. R. Ghodssi and
P. Lin, Springer US, Boston, MA, 2011, vol. 1.
7 Spin Crossover in Transition Metal Compounds, Top. Curr.
Chem., ed. P. Gütlich and H. A. Goodwin, Springer, Berlin,
Heidelberg, 2004, vol. 233–235.
8 A. Bousseksou, G. Molnár, L. Salmon and W. Nicolazzi,
Chem. Soc. Rev., 2011, 40, 3313–3335.
9 P. Guionneau, Dalton Trans., 2014, 43, 382–393.
10 S. Cobo, D. Ostrovskii, S. Bonhommeau, L. Vendier,
G. Molnár, L. Salmon, K. Tanaka and A. Bousseksou, J. Am.
Chem. Soc., 2008, 130, 9019–9024.
11 H. J. Shepherd, I. A. Gural'skiy, C. M. Quintero, S. Tricard,
L. Salmon, G. Molnár and A. Bousseksou, Nat. Commun.,
2013, 4, 2607.
12 R. Cabrera, E. Merced, N. Sepúlveda and F. E. Fernández, J.
Appl. Phys., 2011, 110, 094510.
13 J. Cao, W. Fan, Q. Zhou, E. Sheu, A. Liu, C. Barrett and J. Wu,
J. Appl. Phys., 2010, 108, 083538.
14 K. Liu, C. Cheng, Z. Cheng, K. Wang, R. Ramesh and J. Wu,
Nano Lett., 2012, 12, 6302–6308.
15 A. Rúa, F. E. Fernández and N. Sepúlveda, J. Appl. Phys., 2010,
107, 074506.
16 K. Wang, C. Cheng, E. Cardona, J. Guan, K. Liu and J. Wu,
ACS Nano, 2013, 7, 2266–2272.
17 C. Faulmann, J. Chahine, I. Malfant, D. de Caro, B. Cormary
and L. Valade, Dalton Trans., 2011, 40, 2480–2485.
18 S. Titos-Padilla, J. M. Herrera, X.-W. Chen, J. J. Delgado and
E. Colacio, Angew. Chem., Int. Ed., 2011, 50, 3290–3293.
19 P. Durand, S. Pillet, E.-E. Bendeif, C. Carteret, M. Bouazaoui,
H. El Hamzaoui, B. Capoen, L. Salmon, S. Hebert,
J. Ghanbaja, L. Arandah and D. Schaniel, J. Mater. Chem.
C, 2013, 1, 1933–1942.
20 S.-W. Lee, J.-W. Lee, S.-H. Jeong, I.-W. Park, Y.-M. Kim and
J.-I. Jin, Synth. Met., 2004, 142, 243–249.
21 A. Nakamoto, Y. Ono, N. Kojima, D. Matsumura and
T. Yokoyama, Chem. Lett., 2003, 32, 336–337.
22 M.-L. Boillot, S. Pillet, A. Tissot, E. Rivière, N. Claiser and
C. Lecomte, Inorg. Chem., 2009, 48, 4729–4736.
23 M. Rubio, R. Hernandez, A. Nogales, A. Roig and D. Lopez,
Eur. Polym. J., 2011, 47, 52–60.
24 J. Larionova, L. Salmon, Y. Guari, A. Tokarev, K. Molvinger,
G. Molnár and A. Bousseksou, Angew. Chem., Int. Ed., 2008,
47, 8236–8240.
25 W. Hellel, A. Ould-Hamouda, J. Degert, J.-F. Létard and
E. Freysz, Appl. Phys. Lett., 2013, 103, 143304.
26 C. Liu, Adv. Mater., 2007, 19, 3783–3790.
27 J. Kröber, J.-P. Audiere, R. Claude, E. Codjovi, O. Kahn,
J. G. Haasnoot, F. Groliere, C. Jay and A. Bousseksou,
Chem. Mater., 1994, 6, 1404–1412.
28 A. Grosjean, P. Négrier, P. Bordet, C. Etrillard, D. Mondieig,
S. Pechev, E. Lebraud, J.-F. Létard and P. Guionneau, Eur. J.
Inorg. Chem., 2013, 796–802.
29 N. O. Moussa, D. Ostrovskii, V. M. Garcia, G. Molnár,
K. Tanaka, A. B. Gaspar, J. A. Real and A. Bousseksou,
Chem. Phys. Lett., 2009, 477, 156–159.
30 V. Q. Nguyen, A. S. Ahmed and R. V. Ramanujan, Adv. Mater.,
2012, 24, 4041–4054.
31 C. Bartual-Murgui, A. Akou, L. Salmon, G. Molnár, C. Thibault,
J. A. Real and A. Bousseksou, Small, 2011, 7, 3385–3391.